1. Technical Field
The present invention relates to a bi-lingual system for automatically converting a character stream of one language into a character stream of another language, and more particularly, to an automatic character converting system in a bi-lingual word processor using a list method and a pronunciation method to obtain highly accurate conversion of one language into another language.
2. Related Art
Computer systems have become increasingly sophisticated and powerful as a result of increases in processing speed and memory capacity. More attention has been given to using computer systems, and particularly, word processors to convert written words from one language to another. Contemporary efforts to translate one language to another are disclosed, for example, U.S. Pat. No. 4,484,305 for Phonetic Multilingual Word Processor issued to Ho, U.S. Pat. No. 4,443,856 for Electronic Translator For Modifying And Speaking Out Sentence issued to Hashimoto et al., U.S. Pat. No. 4,654,798 for System Of Simultaneous Translation Into A Plurality Of Languages With Sentence Forming Capabilities issued to Taki et al., U.S. Pat. No. 5,091,950 for Arabic Language Translating Device With Pronunciation Capability Using Language Pronunciation Rules issued to Ahmed, U.S. Pat. No. 5,127,748 for Documentation System Having Multilingual Function issued to Okimoto et al., U.S. Pat. No. 5,384,701 for Language Translation System issued to Stentiford et al., U.S. Pat. No. 5,477,451 for Method And System For Natural Language Translation issued to Brown et al., U.S. Pat. No. 5,623,682 for Apparatus For Processing Documentary Information issued to Fukunaga.
A computer system for executing word processing typically has data inputting, data processing, and data outputting capabilities. Word processors consist a variety of capabilities, such as automatic character converting and spell checking. Generally, automatic character conversion relates to a process for converting a character stream of one language into that of another language while working with a bilingual word processor. Although a bilingual conversion system may be used to convert any combination of two languages, for the purposes of this discussion, the combination of Korean/English languages is described.
First of all the automatic conversion method is used to convert an unrecognizable character stream of one language into another language. When a stream of character is input to form a word, the word should have a proper character stream such that an initial sound should precede a middle sound which should be followed by a final sound. Consequently, when an incorrect order of sounds is detected, the automatic conversion system automatically switches the word into other language (Korean to English or vice versa). Another method is to allow the user to define the placement of the automatic conversion. Therefore, when an unrecognizable character stream of one language is input, the user defines the placement of the unrecognizable character stream to be converted automatically into another language. Since such a method can be controlled by the user, the accuracy of conversion tends to be very high. However, I have observed that both methods have drawbacks. Since the automata method converts words based on a recognizable character stream, when a bi-lingual word or a character stream that has consecutive vowels or consonants is input, the character conversion cannot be accurately performed, particularly when the consecutive words, consonants or vowels are escaped from the combination rule. Specifically, when a consecutive vowels or consonants is input, such character stream will be converted into another language, even though such character stream may signify something else.